Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    named after, type of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    のもの
    0.86
    Things
    0.77
     things
    0.72
    avoir
    0.72
    thing
    0.71
     coisas
    0.70
     THINGS
    0.69
    things
    0.69
    ım
    0.69
    どこ
    0.67
    POSITIVE LOGITS
     नामक
    1.15
     নামক
    1.09
     sebuah
    1.08
     ovvero
    1.08
    と呼ばれる
    1.03
    一種
    1.02
     czyli
    1.01
    一种
    1.00
     μια
    0.99
     எனும்
    0.97
    Act Density 0.108%

    No Known Activations